{"id":{"repo_id":"clemson","oai_identifier":"oai:open.clemson.edu:all_theses-5696"},"canonical_url":"https://search.ndltd.org/etd/clemson/oai:open.clemson.edu:all_theses-5696","repository":{"repo_id":"clemson","name":"Clemson University","base_url":"https://open.clemson.edu/do/oai/"},"display":{"title":"Understanding Policy Transfer Under Decomposition and Coordination: Sequential Reinforcement Learning With Relaxation-Based Feasibility Control","abstract":"This thesis investigates how decomposition and coordination (D&C) choices influence exploration, learning, and transfer in reinforcement learning (RL) based design frameworks. A unified methodology is developed to evaluate generality under decomposition and to demonstrate sequential coordination on a representative physical task. The generality study holds the underlying physics constant while varying problem presentation - specifically scalarization weights, objective orientations, and feasibility handling. A train-evaluate matrix is constructed in which each trained policy is assessed across all alternative decompositions. Set-based diagnostics quantify generality through exact-match identity, Jaccard similarity of distinct feasible states, total feasible coverage, and Pareto-front quality. Results show that different decompositions can yield comparable hypervolume yet diverge in identity, revealing that presentation alters exploration paths. Adjusting or reorienting select objectives improves alignment in some cases, indicating that presentation grammar and objective pairing are critical coordination and design choices. Building on this foundation, a sequential demonstration applies the framework to Mars terrain navigation using MOLA data. A relaxation variable ρ is introduced as an explicit coordination channel controlling permissible deviations from slope constraints. Deterministic evaluations across increasing ρmax values show a transition from stalled trajectories at ρmax = 0 to efficient motion with higher reach and modest hazard at moderate ρmax. The behavior exhibits consistent temporal patterns of relaxation and recovery, indicating structured coordination rather than memorization. The research contributes: (1) a reproducible protocol for assessing presentation sensitivity via a train- evaluate matrix and complementary set metrics; (2) a constructive mapping from D&C constructs to RL and constrained-MDP elements that clarifies how presentation induces distributional shifts; and (3) a sequential relaxation mechanism that operationalizes coordination within RL while yielding planner- usable option sets. These results advance understanding of how problem formulation governs learning behavior and establish a basis for coordination-aware, presentation-robust decision frameworks.","abstract_html":"This thesis investigates how decomposition and coordination (D&amp;C) choices influence exploration, learning, and transfer in reinforcement learning (RL) based design frameworks. A unified methodology is developed to evaluate generality under decomposition and to demonstrate sequential coordination on a representative physical task. The generality study holds the underlying physics constant while varying problem presentation - specifically scalarization weights, objective orientations, and feasibility handling. A train-evaluate matrix is constructed in which each trained policy is assessed across all alternative decompositions. Set-based diagnostics quantify generality through exact-match identity, Jaccard similarity of distinct feasible states, total feasible coverage, and Pareto-front quality. Results show that different decompositions can yield comparable hypervolume yet diverge in identity, revealing that presentation alters exploration paths. Adjusting or reorienting select objectives improves alignment in some cases, indicating that presentation grammar and objective pairing are critical coordination and design choices. Building on this foundation, a sequential demonstration applies the framework to Mars terrain navigation using MOLA data. A relaxation variable ρ is introduced as an explicit coordination channel controlling permissible deviations from slope constraints. Deterministic evaluations across increasing ρmax values show a transition from stalled trajectories at ρmax = 0 to efficient motion with higher reach and modest hazard at moderate ρmax. The behavior exhibits consistent temporal patterns of relaxation and recovery, indicating structured coordination rather than memorization. The research contributes: (1) a reproducible protocol for assessing presentation sensitivity via a train- evaluate matrix and complementary set metrics; (2) a constructive mapping from D&amp;C constructs to RL and constrained-MDP elements that clarifies how presentation induces distributional shifts; and (3) a sequential relaxation mechanism that operationalizes coordination within RL while yielding planner- usable option sets. These results advance understanding of how problem formulation governs learning behavior and establish a basis for coordination-aware, presentation-robust decision frameworks.","abstract_has_math":false,"creators":["Lal, Aannand"],"institution":null,"degree_name":"Master of Science (MS)","degree_level":"Thesis","degree_discipline":"Mechanical Engineering","degree_department":null,"school":null,"contributors":["Dr. Cameron Turner","Dr. John Wagner","Dr. Laura Redmond"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T07:33:56Z","subjects":["decomposition and coordination; multiobjective optimization; reinforcement learning; constrained MDPs; presentation bias; Mars navigation; sequential relaxation","Mechanical Engineering","Navigation, Guidance, Control and Dynamics","Systems Engineering and Multidisciplinary Design Optimization"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://open.clemson.edu/all_theses/4675","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://open.clemson.edu/all_theses/4675","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.clemson.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Aopen.clemson.edu%3Aall_theses-5696","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Dr. Cameron Turner","Dr. John Wagner","Dr. Laura Redmond"]},{"key":"dc:creator","label":"Author","values":["Lal, Aannand"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2025-11-26T08:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2025-12-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Mechanical Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science (MS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["decomposition and coordination; multiobjective optimization; reinforcement learning; constrained MDPs; presentation bias; Mars navigation; sequential relaxation","Mechanical Engineering","Navigation, Guidance, Control and Dynamics","Systems Engineering and Multidisciplinary Design Optimization"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://open.clemson.edu/all_theses/4675"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis investigates how decomposition and coordination (D&C) choices influence exploration, learning, and transfer in reinforcement learning (RL) based design frameworks. A unified methodology is developed to evaluate generality under decomposition and to demonstrate sequential coordination on a representative physical task. The generality study holds the underlying physics constant while varying problem presentation - specifically scalarization weights, objective orientations, and feasibility handling. A train-evaluate matrix is constructed in which each trained policy is assessed across all alternative decompositions. Set-based diagnostics quantify generality through exact-match identity, Jaccard similarity of distinct feasible states, total feasible coverage, and Pareto-front quality. Results show that different decompositions can yield comparable hypervolume yet diverge in identity, revealing that presentation alters exploration paths. Adjusting or reorienting select objectives improves alignment in some cases, indicating that presentation grammar and objective pairing are critical coordination and design choices. Building on this foundation, a sequential demonstration applies the framework to Mars terrain navigation using MOLA data. A relaxation variable ρ is introduced as an explicit coordination channel controlling permissible deviations from slope constraints. Deterministic evaluations across increasing ρmax values show a transition from stalled trajectories at ρmax = 0 to efficient motion with higher reach and modest hazard at moderate ρmax. The behavior exhibits consistent temporal patterns of relaxation and recovery, indicating structured coordination rather than memorization. The research contributes: (1) a reproducible protocol for assessing presentation sensitivity via a train- evaluate matrix and complementary set metrics; (2) a constructive mapping from D&C constructs to RL and constrained-MDP elements that clarifies how presentation induces distributional shifts; and (3) a sequential relaxation mechanism that operationalizes coordination within RL while yielding planner- usable option sets. These results advance understanding of how problem formulation governs learning behavior and establish a basis for coordination-aware, presentation-robust decision frameworks."]},{"key":"dc:title","label":"Title","values":["Understanding Policy Transfer Under Decomposition and Coordination: Sequential Reinforcement Learning With Relaxation-Based Feasibility Control"]}]}],"canonical_facts":{"dc:contributor":["Dr. Cameron Turner","Dr. John Wagner","Dr. Laura Redmond"],"dc:creator":["Lal, Aannand"],"dc:date.available":["2025-11-26T08:00:00Z"],"dc:date.created":["2025-12-01T08:00:00Z"],"dc:description.abstract":["This thesis investigates how decomposition and coordination (D&C) choices influence exploration, learning, and transfer in reinforcement learning (RL) based design frameworks. A unified methodology is developed to evaluate generality under decomposition and to demonstrate sequential coordination on a representative physical task. The generality study holds the underlying physics constant while varying problem presentation - specifically scalarization weights, objective orientations, and feasibility handling. A train-evaluate matrix is constructed in which each trained policy is assessed across all alternative decompositions. Set-based diagnostics quantify generality through exact-match identity, Jaccard similarity of distinct feasible states, total feasible coverage, and Pareto-front quality. Results show that different decompositions can yield comparable hypervolume yet diverge in identity, revealing that presentation alters exploration paths. Adjusting or reorienting select objectives improves alignment in some cases, indicating that presentation grammar and objective pairing are critical coordination and design choices. Building on this foundation, a sequential demonstration applies the framework to Mars terrain navigation using MOLA data. A relaxation variable ρ is introduced as an explicit coordination channel controlling permissible deviations from slope constraints. Deterministic evaluations across increasing ρmax values show a transition from stalled trajectories at ρmax = 0 to efficient motion with higher reach and modest hazard at moderate ρmax. The behavior exhibits consistent temporal patterns of relaxation and recovery, indicating structured coordination rather than memorization. The research contributes: (1) a reproducible protocol for assessing presentation sensitivity via a train- evaluate matrix and complementary set metrics; (2) a constructive mapping from D&C constructs to RL and constrained-MDP elements that clarifies how presentation induces distributional shifts; and (3) a sequential relaxation mechanism that operationalizes coordination within RL while yielding planner- usable option sets. These results advance understanding of how problem formulation governs learning behavior and establish a basis for coordination-aware, presentation-robust decision frameworks."],"dc:identifier":["https://open.clemson.edu/all_theses/4675"],"dc:subject":["decomposition and coordination; multiobjective optimization; reinforcement learning; constrained MDPs; presentation bias; Mars navigation; sequential relaxation","Mechanical Engineering","Navigation, Guidance, Control and Dynamics","Systems Engineering and Multidisciplinary Design Optimization"],"dc:title":["Understanding Policy Transfer Under Decomposition and Coordination: Sequential Reinforcement Learning With Relaxation-Based Feasibility Control"],"thesis:degree_discipline":["Mechanical Engineering"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["Master of Science (MS)"]},"updated_at":"2026-11-19T07:33:56Z"}